Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

connection to ISAAC server seems to be good, but no visualization #4870

Open
onezillion opened this issue Apr 10, 2024 · 23 comments
Open

connection to ISAAC server seems to be good, but no visualization #4870

onezillion opened this issue Apr 10, 2024 · 23 comments
Assignees
Labels

Comments

@onezillion
Copy link

onezillion commented Apr 10, 2024

Dear all,

I am trying to use PIConGPU with ISAAC visualization. My current status is very similar to issue #3410 (part of).
The installation is mostly via spack, and isaac & isaac-server are installed with @develop (1.6.0). Anyway, installation went smoothly.
PIConGPU is on dev branch with tag of 0.7.0-530-g71ec38322 (though it shows 0.8.0-dev during simulation).
The simulation is running on a linux VM with ubuntu 22.04 and with 2x4 GPU cores.

The current situation is, the simulation runs smoothly (myLWFA example, with 4.cfg or 4_isaac.cfg). The png figures looks good; openpmd data looks good; stdout of compiling and of the simulation looks ok.

During compiling I can see:
...
-- ISAAC_RENDERER_OPTIMIZED is enabled
-- Found ISAAC: /home/u07khl00/spack/opt/spack/linux-ubuntu22.04-skylake_avx512/gcc-12.3.0/isaac-develop-52jjftkdf6t3sxsio3ahd2c5sbdjftys/lib/cmake/ISAAC

Optional Dependencies:
openPMD: ON
PNGwriter: ON
ISAAC: ON
...

During the beginning of the simulation, I can see:
...
PIConGPUVerbose INPUT_OUTPUT(32) | ISAAC Init succeded
...

The server is running on the same machine (localhost) in the background. And when the simulation starts (isaac connect to localhost), from the server stdout, I can see:
...
New connection, giving id 1 (control)
Group complete, sending to connected interfaces
...

I am using the tunnel via putty (as mentioned in isaac: tunnel.md), and I open isaac\client\html\visualisation.html.
I click on the observe button and go into the observing page. The message of server stdout shows
...
URIImageConnector: Openend URI JPEG Stream
...

Everything looks find except there is no visualization of the simulation (fields, particles, etc.).

image

image

The particle num is correct, the cell num is correct, the time step shows good correspondence to simulation stdout. The dimension is also correct. I also tried clicking the "update minmax interval" button repeatedly, and the max/min values are updated repeatedly with reasonable values.

image

I tried running simulation and server on two different VMs, but still got no visualization.

In the earlier post #3410, there is no mentioning on how to solve this. It was just fixed somehow.
I'm here seeking help to solve this issue. Thank you in advance.

@psychocoderHPC
Copy link
Member

Hi @onezillion thanks for opening the issue.
Have you activated the data source with the slider so that the B field, E and J will become visible?

I need to retest if we broke something with the current dev branch. We switched with PIConGPU dev fully to native alpaka instead of using the cupla library.
If the slider will not solve the issue I will have a look into it, could you report if the slider start showing something?

@psychocoderHPC
Copy link
Member

@PrometheusPi We need to test ISAAC with the latest dev branch.

@PrometheusPi
Copy link
Member

@psychocoderHPC Thanks for pinging - I will have a look

@onezillion
Copy link
Author

Hi @onezillion thanks for opening the issue. Have you activated the data source with the slider so that the B field, E and J will become visible?

I need to retest if we broke something with the current dev branch. We switched with PIConGPU dev fully to native alpaka instead of using the cupla library. If the slider will not solve the issue I will have a look into it, could you report if the slider start showing something?

Thanks for your reply.
I am sorry for my dumb question. What is a slider? How can I interact with it or activate it?
I am attaching the stdout during compiling for your reference.

myLWFA_compile.txt

and in isaac.param, I use
...
using Particle_Seq = VectorAllSpecies;
using Native_Seq = MakeSeq_t<FieldE, FieldB, FieldJ>;
using Density_Seq = deriveField::CreateEligible_t<Particle_Seq, deriveField::derivedAttributes::Density>;
using Fields_Seq = MakeSeq_t<Native_Seq, Density_Seq>;
using VectorFields_Seq = Native_Seq;
...

I use 4_isaac.cfg without modification

Thanks again for your reply.

@psychocoderHPC
Copy link
Member

psychocoderHPC commented Apr 16, 2024

Screenshot from 2021-08-04 16-59-39

I mean the slider in the gui under "Volume Weights and Partticle Sizes."
Please do not forget to klick on the update button when you change something.

@PrometheusPi
Copy link
Member

Hi @onezillion,
I tried to reproduce your issue (using LaserWakfield and 4_isaac.cfg) and could not.
I compiled ISAAC with PIConGPU's latest dev (yes, it says PIConGPU: 0.8.0-dev at startup)

grafik

The only issue I had was that I had to use a previously compiled server. However, that software did not change for years (that's why I had issues building it.)

From the log files you shared, I see a few strange things:

  • your server states:
New connection, giving id 1 (control)
Group complete, sending to connected interfaces

but your image is only showing simulation ID 0. Do you run two simulations?

  • Do you see the connection from your bwoser in the server log, as:
ISAAC Connection from <your IP address> (<your IP address>)!
callback_isaac: LWS_CALLBACK_ESTABLISHED
URIImageConnector: Openend URI JPEG Stream
callback_isaac: LWS_CALLBACK_FILTER_PROTOCOL_CONNECTION

Did you adjust any of the 3 ports that are set by default or did the server or client needed to reroute, e.g. the web port? (because data transfer seems to work)

@PrometheusPi
Copy link
Member

PS: did you activate moving window? Because your Dimensions look different than the default.

@onezillion
Copy link
Author

onezillion commented Apr 17, 2024

Thank you for your replies. Sorry that I am away for a business trip.
The ID number difference may be due to that the screenshot and stdout are obtained from different simulation runs. But each time, only one simulation is running. I will confirm this later and check the ID number.

I do remember that similar messages have occurred in server stdout:
ISAAC Connection from ()!
callback_isaac: LWS_CALLBACK_ESTABLISHED
……
I will also double confirm this and make screenshot later.

As for the slider, I have been trying changing the max/min limits and playing around the color map before I posted this issue (I did click the update button). I will also check this again.

I have tried both modified and default port settings. The tunelling settings are as suggested by the isaac page. I think there is no firewall between the vm machine and my pc terminal, or at least the default ports are not blocked.

I cannot remember the moving window setting, but I did not change this setting.

Sorry that I am still away for a business trip until this weekend. I will check these suggestions on this weekend.

@onezillion
Copy link
Author

onezillion commented Apr 20, 2024

I am attaching the stdout and screenshots throughout the whole procedure from loading spack modules to establishing the visualization in the following.
I use putty and tunnel connection (localhost)
The issue persists. It is really strange. Maybe it is a machine dependent problem? Is there any system-based setting that I should take care of?

PIConGPU startup

load spack modules

$> source picongpu.profile
export env vars & load pkgs for PIConGPU ......
-- linux-ubuntu22.04-skylake_avx512 / [email protected] ----------------
[email protected]  [email protected]  [email protected]  glm@develop  [email protected]  [email protected]  isaac@develop  isaac-server@develop  [email protected]  [email protected]  [email protected]  [email protected]  [email protected]  [email protected]  [email protected]
==> 15 loaded packages

load isaac_server & put it into background

The port settings are all default

$> source isaac_server.profile
Using web_port=2459, tcp_port=2458 and sim_port=2460

lwsts[2130989]: Initial logging level 1031
lwsts[2130989]: Libwebsockets version: 2.2.0 u07khl00@pca1-
lwsts[2130989]: IPV6 not compiled in
lwsts[2130989]: libev support not compiled in
lwsts[2130989]: libuv support not compiled in
lwsts[2130989]:  Threads: 1 each 1024 fds
lwsts[2130989]:  mem: platform fd map:  8192 bytes
lwsts[2130989]:  Compiled with OpenSSL support
lwsts[2130989]:  SSL disabled: no LWS_SERVER_OPTION_DO_SSL_GLOBAL_INIT
lwsts[2130989]: Creating Vhost 'default' port 2459, 2 protocols, IPv6 off
lwsts[2130989]:  Listening on port 2459
lwsts[2130989]:  mem: per-conn:          568 bytes + protocol rx buf
lwsts[2130989]:  canonical_hostname = pca1
Running ISAAC Master
Starting insitu plugin listener
Launching WebSocketDataConnector
Launching TCPDataConnector
Launching JPEG_URI_Stream
u07khl00    2136  0.0  0.0  17312 10032 ?        Ss   Apr08   0:00 /lib/systemd/systemd --user
u07khl00    2137  0.0  0.0 170568  4656 ?        S    Apr08   0:00 (sd-pam)
u07khl00  835518  0.0  0.0  81400   780 ?        Ss   Apr11   0:00 gpg-agent --homedir /home/u07khl00/spack/opt/spack/gpg --use-standard-socket --daemon
root      964349  0.0  0.0  17788 10808 ?        Ss   Apr12   0:00 sshd: u07khl00 [priv]
u07khl00  964447  0.0  0.0  34428 24588 ?        S    Apr12   3:34 sshd: u07khl00@notty
u07khl00  964448  0.0  0.0   8900  5068 ?        Ss   Apr12   0:03 -bash
root     2077953  0.0  0.0  17784 11032 ?        Ss   Apr18   0:00 sshd: u07khl00 [priv]
u07khl00 2078051  0.0  0.0  34504 24556 ?        S    Apr18   2:44 sshd: u07khl00@notty
u07khl00 2078052  0.0  0.0   8900  5148 ?        Ss   Apr18   0:00 -bash
root     2130523  0.0  0.0  17792 10872 ?        Ss   04:47   0:00 sshd: u07khl00 [priv]
u07khl00 2130624  0.0  0.0  17792  7908 ?        S    04:47   0:00 sshd: u07khl00@notty
u07khl00 2130627  0.0  0.0   8900  5128 ?        Ss   04:47   0:00 -bash
root     2130828  0.0  0.0  17784 10916 ?        Ss   04:50   0:00 sshd: u07khl00 [priv]
u07khl00 2130887  0.0  0.0  17784  7916 ?        S    04:51   0:00 sshd: u07khl00@pts/0
u07khl00 2130888  0.0  0.0   9936  6700 pts/0    Ss   04:51   0:00 -bash
u07khl00 2130989  0.0  0.0 178480  5340 pts/0    Sl   04:52   0:00 /home//u07khl00/spack/opt/spack/linux-ubuntu22.04-skylake_avx512/gcc-12.3.0/isaac-server-develop-6fqx5sc3wayrsipstou6utlufvhi2hmn/bin/isaac
u07khl00 2130990  0.0  0.0  10312  2488 pts/0    R+   04:52   0:00 ps aux
u07khl00 2130991  0.0  0.0   6636  2308 pts/0    S+   04:52   0:00 grep --color=auto u07khl00

use client "visualisation.html" to connect to isaac server; server stdout:

the message callback_isaac: LWS_CALLBACK_ESTABLISHED is observed.

callback_isaac: LWS_CALLBACK_FILTER_PROTOCOL_CONNECTION
ISAAC Connection from pca1 (172.21.31.1)!
callback_isaac: LWS_CALLBACK_ESTABLISHED

the web page screenshot
image

I use default example, LWFA, and modify isaac.param

#pragma once
#include "picongpu/particles/particleToGrid/derivedAttributes/DerivedAttributes.def"
namespace picongpu
{
    namespace isaacP
    {
        namespace deriveField = particles::particleToGrid;
        using Particle_Seq = VectorAllSpecies;
        using Native_Seq = MakeSeq_t<FieldE, FieldB, FieldJ>;
        using Density_Seq = deriveField::CreateEligible_t<Particle_Seq, deriveField::derivedAttributes::Density>;
        using Fields_Seq = MakeSeq_t<Native_Seq, Density_Seq>;
        using VectorFields_Seq = Native_Seq;
    } // namespace isaacP
} // namespace picongpu

use 4_isaac.cfg

TBG_wallTime="1:00:00"
TBG_devices_x=1
TBG_devices_y=4
TBG_devices_z=1

TBG_gridSize="192 2048 192"
TBG_steps="2048"

TBG_restartLoop="--checkpoint.restart.loop 10000"
TBG_isaac="--isaac.period 1"
TBG_plugins="!TBG_isaac"
TBG_deviceDist="!TBG_devices_x !TBG_devices_y !TBG_devices_z"
TBG_programParams="-d !TBG_deviceDist \
                   -g !TBG_gridSize   \
                   -s !TBG_steps      \
                   !TBG_restartLoop  \
                   !TBG_plugins      \
                   --versionOnce"
TBG_tasks="$(( TBG_devices_x * TBG_devices_y * TBG_devices_z ))"
"$TBG_cfgPath"/submitAction.sh

build PIConGPU

pic-build -c"-DPIC_USE_ISAAC=ON -DPIC_USE_openPMD=ON -DPIC_USE_PNGwriter=ON"
The stdout is pretty lengthy, so I attach it as a file
PIConGPU compilation.txt

run PIConGPU

I found that the moving window is OFF by default
tbg -s bash -c etc/picongpu/4_isaac.cfg -t etc/picongpu/bash/mpiexec.tpl $SCRATCH/runs/myLWFA_test1

this time, the ID numbers are the same in the terminal stdout, New connection, giving id 0 (control), and in the web page (screenshot in the following)

$> tbg -s bash -c etc/picongpu/4_isaac.cfg -t etc/picongpu/bash/mpiexec.tpl $SCRATCH/runs/myLWFA_test1
Running program...
export env vars & load pkgs for PIConGPU ......
-- linux-ubuntu22.04-skylake_avx512 / [email protected] ----------------
[email protected]  [email protected]  [email protected]  glm@develop  [email protected]  [email protected]  isaac@develop  isaac-server@develop  [email protected]  [email protected]  [email protected]  [email protected]  [email protected]  [email protected]  [email protected]
==> 15 loaded packages
PIConGPU: 0.8.0-dev
  Build-Type: Release

Third party:
  OS:         Linux-5.15.0-101-generic
  arch:       x86_64
  CXX:        GNU (12.3.0)
  CMake:      3.27.7
  CUDA:       12.2.128
  mallocMC:   2.6.0
  Boost:      1.83.0
  MPI:
    standard: 3.1
    flavor:   OpenMPI (4.1.6)
  PNGwriter:  0.7.0
  openPMD:    0.15.2
PIConGPUVerbose PHYSICS(1) | Sliding Window is OFF
1: -1
2: -1
3: 1
4: -1
5: -1
6: -1
7: -1
8: -1
9: -1
10: -1
11: -1
12: -1
13: -1
14: -1
15: -1
16: -1
17: -1
18: -1
19: -1
20: -1
21: -1
22: -1
23: -1
24: -1
25: -1
26: -1
New connection, giving id 0 (control)
Group complete, sending to connected interfaces
PIConGPUVerbose PHYSICS(1) | used Random Number Generator: RNGProvider3XorMin seed: 42
PIConGPUVerbose PHYSICS(1) | Field solver condition: c * dt <= 1.00229 ? (c * dt = 1)
PIConGPUVerbose PHYSICS(1) | Incident field "GaussianPulse_with_GaussianPulse" numerical dispersion: v_phase = 0.999412 * c
PIConGPUVerbose PHYSICS(1) | Resolving plasma oscillations?
   Estimates are based on DensityRatio to BASE_DENSITY of each species
   (see: density.param, speciesDefinition.param).
   It does not cover other forms of initialization
PIConGPUVerbose PHYSICS(1) | species e: omega_p * dt <= 0.1 ? (omega_p * dt = 0.0247974)
PIConGPUVerbose PHYSICS(1) | macro particles per device: 37748736
PIConGPUVerbose PHYSICS(1) | typical macro particle weighting: 6955.06
PIConGPUVerbose PHYSICS(1) | UNIT_SPEED 2.99792e+08
PIConGPUVerbose PHYSICS(1) | UNIT_TIME 1.39e-16
PIConGPUVerbose PHYSICS(1) | UNIT_LENGTH 4.16712e-08
PIConGPUVerbose PHYSICS(1) | UNIT_MASS 6.33563e-27
PIConGPUVerbose PHYSICS(1) | UNIT_CHARGE 1.11432e-15
PIConGPUVerbose PHYSICS(1) | UNIT_EFIELD 1.22627e+13
PIConGPUVerbose PHYSICS(1) | UNIT_BFIELD 40903.8
PIConGPUVerbose PHYSICS(1) | UNIT_ENERGY 5.69418e-10
PIConGPUVerbose PHYSICS(1) | Resolving Debye length for species "e"?
PIConGPUVerbose PHYSICS(1) | Estimate used momentum variance in 288000 supercells with at least 10 macroparticles each
PIConGPUVerbose PHYSICS(1) | 288000 (100 %) supercells had local Debye length estimate not resolved by a single cell
PIConGPUVerbose PHYSICS(1) | Estimated weighted average temperature 0 keV and corresponding Debye length 0 m.
   The grid has 0 cells per average Debye length
initialization time: 20sec 655msec = 20.655 sec
  0 % =        0 | time elapsed:                   28msec | avg time per step:   0msec
  4 % =      102 | time elapsed:             4sec  62msec | avg time per step:  23msec
  9 % =      204 | time elapsed:             8sec 111msec | avg time per step:  23msec
 14 % =      306 | time elapsed:            12sec 250msec | avg time per step:  24msec
 19 % =      408 | time elapsed:            16sec 590msec | avg time per step:  25msec

the web page becomes
image

use the "observe" button to see the visualization

a message showed up in the terminal URIImageConnector: Openend URI JPEG Stream

I tried the max/min interval button
I played around the sliders, and of course pushed the submit/update button
the web page screenshots
image

image

image

try different slider/colormap settings

just found that I pick the wrong weight option. will redo this test and update later.

image

image

image

@psychocoderHPC
Copy link
Member

I assume the problem is that you installed ISAAC with spack. Due to time constrains we stopped maintaining the spack packages for ISAAC.
You can compile all dependencies with spack but currently you should compile ISAAC by hand.

You can grep all required commands from our CI job:

export GLM_ROOT=/opt/glm/0.9.9.9-dev
export CMAKE_PREFIX_PATH=$GLM_ROOT:$CMAKE_PREFIX_PATH
git clone https://github.com/g-truc/glm.git
cd glm
git checkout 6ad79aae3eb5bf809c30bf1168171e9e55857e45
mkdir build
cd build
cmake ../ -DCMAKE_INSTALL_PREFIX=$GLM_ROOT -DGLM_TEST_ENABLE=OFF
make install
cd $CI_PROJECT_DIR
git clone https://github.com/ComputationalRadiationPhysics/isaac.git
cd isaac
git checkout 195420ca1c43f6148c7f6c3a10ad9cad32e04d6a
mkdir build_isaac
cd build_isaac
cmake ../lib/ -DCMAKE_INSTALL_PREFIX=$ISAAC_ROOT
make install
cd ..

And for the server the install description of ISAAC should work fine if you provide libwebsockets with spack https://github.com/ComputationalRadiationPhysics/isaac/blob/dev/INSTALL.md#the-server

@PrometheusPi
Copy link
Member

@onezillion Did building the software stack, as @psychocoderHPC recommended, solved our problem?

@onezillion
Copy link
Author

onezillion commented Apr 27, 2024

Hi all,

Sorry for my late response. I recompiled isaac from source for isaac/lib and isaac/server.
All other dependencies are the same as before (spack).
I also git fetch and git pull in picongpu folder.

image
image
image

the pic-build runs smoothly
image
image

When opening the client web page and clicking the observe button, the corresponding server messages also show up
螢幕擷取畫面 2024-04-28 004035
image

But still no visualization:
螢幕擷取畫面 2024-04-28 004232

Tried the sliders, but still no luck:
image

Since this does not occur commonly, I guess it may be related to our VM machine?

@psychocoderHPC
Copy link
Member

Tried the sliders, but still no luck:

Do you klicked on Update weights after changing the sliders?

Since this does not occur commonly, I guess it may be related to our VM machine?

What do you mean by this? Is it sometimes working?

Which browser do you use? I suggest testing firefox and Chrome if you not already tested these.

@psychocoderHPC
Copy link
Member

There is on option for the server what all images get stored as jpg in the server folder.
Maybe you can try to use this option and change sliders in the webinterface and check if the jpg's has non black images.
In that case this will point to a connection/port issue from the server to the web client.

I do not think that the VM is the issue.

@onezillion
Copy link
Author

onezillion commented Apr 29, 2024

Thank you for your reply.

Do you klicked on Update weights after changing the sliders?

Yes, I did click the update button. I even clicked on all the update buttons that I could find in the page

Since this does not occur commonly, I guess it may be related to our VM machine?

What do you mean by this? Is it sometimes working?

No, I mean: since it seems that this issue only happened to me (I do not see similar case being reported except one, which was quickly solved somehow), I guess that maybe certain settings in our VM machine lead to this issue.

Which browser do you use? I suggest testing firefox and Chrome if you not already tested these.

I am using chrome, also tried firefox and edge when I first tried isaac and found no visualization. That was like 2-3 weeks ago. At that time, I also turned on jpeg output, and all I got was just blank & black images. Anyway, let me try this again with manually compiled isaac later.

Actually, my very first try was with only jpeg output and without server connection and found all are just blank & black images. Then I tried the server connection. Then you know the results.

@onezillion
Copy link
Author

onezillion commented May 6, 2024

Dear all,

I tried jpeg output from isaac server but no luck. Here is the screenshot and a jpeg file. I also tried edge, chrome, chrome in privacy and firefox in the screenshot. All other settings are the same as my last shot: isaac from source and other deps from spack. I also did git fetch && git pull for picongpu.

螢幕擷取畫面 2024-05-06 085407

螢幕擷取畫面 2024-05-06 085456

1815

isaac server is loaded by $ISAAC_server_DIR/isaac --dump ~/isaac_plots --url 127.0.0.1 & :

螢幕擷取畫面 2024-05-06 090552

Could it be due to libjpeg-turbo? If that is possible, I will try install libjpeg from source later.

@psychocoderHPC
Copy link
Member

Could it be due to libjpeg-turbo? If that is possible, I will try install libjpeg from source later.

This could be. I suggest try to compile it. I am out of ideas if this is not working :-(

@psychocoderHPC
Copy link
Member

You could also try to change the background color of the simulation, under the simulation image in the web control interface.
This will show if at least the image is transfered correctly or if the rendering is broken.

@onezillion
Copy link
Author

onezillion commented May 7, 2024

Changing background color from control page also changes the jpeg output from isaac server.

image

Does this imply that libjpeg-turbo works well? I totally have no clue now.

Anyway, I will try re-compile libjpeg-turbo from source later.

@psychocoderHPC
Copy link
Member

Does this imply that libjpeg-turbo works well?

I assume because if I remember correctly the background is already rendered into the image on the device.
I will try it asap by my own to see if I get the full chain running by myself.

@psychocoderHPC
Copy link
Member

psychocoderHPC commented May 8, 2024

@onezillion I tried it now by my own on our dev server, here are some snippets how I compiled. I mixed spack and glm compile by hand because I have not tested the latest glm releases.

install

export GLM_ROOT=$HOME/lib/glm/0.9.9.9-dev
export CMAKE_PREFIX_PATH=$GLM_ROOT:$CMAKE_PREFIX_PATH
git clone https://github.com/g-truc/glm.git
cd glm
git checkout 6ad79aae3eb5bf809c30bf1168171e9e55857e45
mkdir build
cd build
cmake ../ -DCMAKE_INSTALL_PREFIX=$GLM_ROOT -DGLM_TEST_ENABLE=OFF
make install
cd ../..

git clone https://github.com/ComputationalRadiationPhysics/isaac.git
cd isaac
mkdir build_isaac
cd build_isaac
export ISAAC_ROOT=$HOME/lib/isaac
export CMAKE_PREFIX_PATH=$ISAAC_ROOT:$CMAKE_PREFIX_PATH

cmake ../lib/ -DCMAKE_INSTALL_PREFIX=$ISAAC_ROOT
make install
cd ../..

spack install  icet
spack install  jansson 
spack install  libjpeg-turbo 
spack install  libwebsockets

compile picongpu

export ISAAC_ROOT=$HOME/lib/isaac
export CMAKE_PREFIX_PATH=$ISAAC_ROOT:$CMAKE_PREFIX_PATH
export GLM_ROOT=$HOME/lib/glm/0.9.9.9-dev
export CMAKE_PREFIX_PATH=$GLM_ROOT:$CMAKE_PREFIX_PATH

# I have problems that icet is not found by cmake therefore I set the prefix path to lib
export CMAKE_PREFIX_PATH=/opt/spack-modules/opt/spack/linux-ubuntu22.04-zen2/gcc-12.2.0/icet-2.1.1-obktzqgpbwaad3qzhpe6xzqxsi3t72ji/lib:$CMAKE_PREFIX_PATH

pic-create ~/workspace/picongpu/share/picongpu/examples/LaserWakefield/ lwfa
cd lwfa
pic-build -b cuda:80

compile server

spack install  jansson 
spack install  libwebsockets
# set openssl include path by hand because of missing include in libwebsockets (I have not investigated the source of the issue)
cmake ../isaac/server/ -DCMAKE_CXX_FLAGS=-I/opt/spack-modules/opt/spack/linux-ubuntu22.04-zen2/gcc-12.2.0/openssl-3.1.3-mqsvskp6tc3zonaqddgvkzmnteaccgm5/include

start server

./isaac 
Using web_port=2459, tcp_port=2458 and sim_port=2460

lwsts[1526522]: Initial logging level 1031
lwsts[1526522]: Libwebsockets version: 2.2.0 rwidera@hal8999-
lwsts[1526522]: IPV6 not compiled in
lwsts[1526522]: libev support not compiled in
lwsts[1526522]: libuv support not compiled in
lwsts[1526522]:  Threads: 1 each 1024 fds
lwsts[1526522]:  mem: platform fd map:  8192 bytes
lwsts[1526522]:  Compiled with OpenSSL support
lwsts[1526522]:  SSL disabled: no LWS_SERVER_OPTION_DO_SSL_GLOBAL_INIT
lwsts[1526522]: Creating Vhost 'default' port 2459, 2 protocols, IPv6 off
lwsts[1526522]:  Listening on port 2459
lwsts[1526522]:  mem: per-conn:          568 bytes + protocol rx buf
lwsts[1526522]:  canonical_hostname = hal8999

start PIConGPU LWFA

./bin/picongpu -d 1 1 1 -g 128 512 128 --periodic 1 0 1 -s 1000 --checkpoint.restart.loop 1000 --isaac.period 1 --isaac.name lwfa

issac

I got images in the client without changing any slider.

@psychocoderHPC
Copy link
Member

@onezillion take care I increased the required glm version for ISAAC to 1.0.0 ComputationalRadiationPhysics/isaac#181

@onezillion
Copy link
Author

Thanks for the reply. Our VM is down for a few days and may be up next week. I will try the suggestions then.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

3 participants